I want to tell you about a sort of creative companion of mine. His name is Dr. Andy J. Pizza (side note: last name not actually Pizza) — and no Ph.D. — and he’s an illustrator. He also hosts a mind-opening podcast about unlocking your creative potential.
Now, during his crazy life transformation, Andy not only went from a jailhouse worker to a now semi-famous podcaster, but, as a prominent artist and creator, Andy understands what he’s good at and markets it exceptionally well. With a little creativity and inspiration, you can change how you view your automation work, too.
Here’s the idea: as Andy’s success shows, any job — including automation work — that requires thinking also requires strategy. For e-commerce teams and marketing technologists, that means understanding how you can glean inspiration, “bend” your martech to help you accomplish your strategy, and refine with time.
Any job, including #automation work that requires thinking also requires strategy CLICK TO TWEET
I shared my thoughts about how to do this — which components of marketing automation you should look for, how you can incorporate them within your strategy, and common obstacles — at Emarsys Revolution 2018 in London.
Check out my full-length 28-minute presentation:
Creating a Marketing Automation Strategy to Work With
Before I explain the different levels of marketing automation work, we need to start with strategy. It’s vital to success that you understand what you need to accomplish (your key objectives) before creating any automations and that you have a good idea of how you want to get there — accounting for the channels your customers use, their receptivity and likelihood to respond to offers, and the frequency at which you should set up each communication.
Here’s the five-step process you need to set in place before doing anything:
- Define Your Goal — Do you want people to make a purchase? Or do you want them to be educated about your brand? Many of us create automations without goals, but without defining a goal, you can’t test and see if you actually accomplish the intention.
- Narrow Down a Starting Point — Do you have the data available to achieve your goal and to get people into that automation? Did a customer make a purchase? Did they abandon their cart? If a new contact who has just joined your email list abandoned a cart, and you want to send them an abandonment message, you have to determine whether you have all the data needed to get them into that automation.
- Look at Your Channels — It’s common, at least in the U.S., to collect more emails than cell phone numbers (vital to sending SMS). Which contact details are important for you: email or cell number? Are they on your site right now? Many people don’t collect more than email addresses — but we need to engage with people based on where they are.
- Ask if You Need to Use Offers — Understand offers. They can encourage purchases, but they aren’t always the best way to do it. If someone has been on your website five out of the last 10 days, has made a ton of purchases, and is back again, they’re probably going to buy anyway and won’t require an offer. But those who are mostly inactive likely DO need an offer to sweeten the deal! Offers aren’t needed for every person in every instance — but you need some sort of filtering to segment your list and automation to essentially intelligently divvy up who gets an offer and where.
- Develop Frequency — Define participation settings. How frequently should a person be included in this automation? Welcome series, for instance, are usually sent only once. But cart abandonment emails may be sent more frequently. Of course, you don’t want to reward customers who abandon their cart every week solely to get 10% off to buy new things.
Once you’ve narrowed down these areas, you’re much more prepared to implement automations that not only work together but which will, collectively, create 1-to-1 customer experiences that will connect back to the strategy you initially set out to achieve.
Now comes the fun part: building and creating the campaigns themselves!
Tools of the Trade: What Automations Do You Need in a Marketing Solution?
Automations can range from the crazy and complex to the smooth and simple. I recommend creating the most concise campaigns possible, and only incorporating as much “stuff” — the minimum viable amount, in fact — as is necessary to meet your goals.
Before we talk about mistakes to avoid, it’s important to start at the foundation of automations. I’m going to break down the three levels of automation (and their purpose) to give you necessary background about how automations all work together as a cohesive, fluid unit.
The three “levels” of automation components to be aware of are:
► Level 1
- Inclusion/Exclusion — Which of our customers will be included in this automation, and which will be excluded?
- Channel Engagement — Will we engage our customer with email, SMS, a Facebook ad, or something else?
A lot of automations look something like this. They’re simple (e.g., if someone is new, they get an email; if not, they’re excluded), which is good. But most e-commerce teams will require more than this.
► Level 2
- Filtering — After a person has been included in an automation, should she continue to be included based on different actions that she takes?
- Scheduling — When should we engage with our customers?
When you add extra layers of strategy, you can take a simple automation to the next level by adding personalization and personality to campaigns. When most people think about automations, this step is what they usually think of: some steps, some wait times, different levels of engagement, filtering people out in different areas, displaying offers depending on history, and more. Then we get into Level 3.
► Level 3
- Automation Linking — Linking gives you flexibility in narrowing down your automation/program blueprint by linking certain steps and attributes. You can link automations and tweak your strategy more regularly by connecting multiple components together.
Level 3 is where you can start getting into what I consider “some really cool stuff.” Level 3 allows you to connect systems together, significantly change levels of engagement, and tweak your strategy.
Where We Go Wrong with Automation
All of these levels of automation architecture are wonderful, but they can be for naught if you neglect to be mindful about the most common mistakes made when designing automations. It’s easier than you might think to make these small errors in the midst of your busy campaign work, and I’ve seen them all made — deliberately and inadvertently — in my current training role and also in my previous role as a campaign designer.
► “A blog told us to do a ____ automation.”
I think we’re all a little guilty of reading a blog and seeing something really, really cool — like “We had 17x ROI just by putting out this ONE automation!” or “We ran a welcome series that gave us a 110% retention rate.” Or “We increased revenue 80% with a cart abandonment series.” It feels like if we just do that one thing, we’re going to be successful, but that’s not always the case.
What works for one brand won’t necessarily work for yours! Every company is different, and every company’s customers are different, too. Don’t buy into the hype of what you read just because it seems innovative. As I’ll note below, you should absolutely take note of unconventional ideas and start to implement them in small doses when and where it makes sense… and constantly iterate for performance.
► “We’re waiting on the email designer.”
Prior to moving to my current training role at Emarsys, as an email designer, I often dealt with people sending me emails they needed edited, updated, or altered. There’s a lot of back and forth — a lot of us have built automations that have sat stagnant for months because we were waiting on one email.
This constant cycle of sending, creating, waiting, editing, waiting, returning, approving… well, it’s enough to drive any email marketer crazy. It’s a major hurdle in the execution of automations, and it shouldn’t be. Be wary of these inefficiencies in your process, and take note of where they’re holding you back.
► The Rube Goldberg Machine approach
A Rube Goldberg Machine is a machine that is way more complex than it needs to be to accomplish one simple task. With these contraptions, there are often too many steps, and it becomes way too complicated.
Many of us are creating automations this way. With astronomically large, complex automations, it’s sometimes hard to narrow down the exact point where something went wrong in order to fix it. We end up building large blueprints to accommodate dozens or even a hundred steps/touchpoints when, if we only looked at the problem anew, we’d see that we could achieve the same end goal by taking a much simpler road.
The 3 most common #marketingautomation mistakes: ‘a blog told us to do it’; waiting on the #email designer, & a Rube Goldberg machine CLICK TO TWEET
Fixing Your Automation Woes
Once you succumb to one or all of these common automation woes, it can be increasingly difficult to stop.
As I’ve always believed, doing things the same way over and over only reinforces the same flawed patterns of thinking. When creating automations, we don’t want to get stuck in that mode of doing, so we must change our way of thinking.
To share a quick anecdote: not too long ago, I worked with a vegan shoe company to build a welcome series with the intention of educating new subscribers and customers. They wanted to execute via email and Facebook, where their audience was most receptive. They also wanted to do a percentage discount instead of a flat dollar amount promotion — and they wanted it to be a one-time offer.
I bring this story up because the company didn’t make things more complicated than they had to be. They wanted to be innovative, but they didn’t deviate and start doing things “just because.” Their process breaks away from the standard way most brands look at welcome programs, but it also aligns nicely with the five steps I shared earlier (goal: educate new subscribers; start: new subscribers; channels: email and Facebook; offer: a custom discount; frequency: one time).
Had this company shot themselves in the foot and given way to common errors in automation execution, the entire thing would have potentially been stuck. Let’s not let that happen to you!
Combatting Common Automation Roadblocks
► “A blog told us to do a ____ automation.”
Consider whether this random idea fits into your goal. Otherwise, you won’t reach your customers in the context within which they need and expect. Evaluate your goals (e.g., to educate people about your brand or to inspire customers to make their first purchase), and then determine if this tactic will help in reaching your customers.
It’s really easy to make an automation, press play, and then to never look at it again. We don’t know if people are actually clicking, we don’t know if they’re buying… we’re kind of like “Well, they got the email. It looks like it’s working.” But we don’t really know if it’s meeting our goal. By having a goal and re-evaluating if you’re meeting your customers where they are, it allows you to move forward.
To address these common issues (especially if you’ve already started implementing ad-hoc tactics like this), here are two tips I recommend:
- Make a copy and edit. If the format of the automation makes sense but it’s not reaching your goals, try making a copy and changing content to fix it.
- Delete it! If this automation doesn’t make sense for your brand, do as Elsa in Frozen does… let it go.
► “We’re waiting on the email designer.”
To avoid this common issue, evaluate:
- What you need to start. Can you launch with only one email? Do you have the data you need to evaluate success?
- What’s the minimum viable product (automation) we need? Create an automation that meets the minimum that is required to be successful, and go from there.
- Start today and link it later. Can you launch with one type of engagement and add additional layers of strategy later?
► The Rube Goldberg Machine approach
Don’t make it too big! Evaluate:
- Where are the problems? Have you received recent complaints of people receiving emails or seeing ads that are not relevant to them?
- Can this be divided into “bite size pieces?” Does this automation need to be contained in one program? Multiple automation branches are a sign that an automation can be split into smaller automations.
- Link our automations. Automations don’t need to stand on their own, so use endpoints to link automations to other starting points.
Following these fixes, suddenly information exchange and knowledge sharing becomes easier, even safeguarded, when people move roles, leave the company, or decide to shift the overarching strategy.
When it comes to what you need in an automation system, there are several key capabilities to consider:
- Inclusion and exclusion rules
- Channel engagement (email, Facebook, SMS, etc.)
- Filtering (moving people around inside of your automation programs)
- Scheduling (emails, for example, to accommodate new products or delays)
- API integrations (linking systems, like Emarsys Web Extend, Magento, and Shopify)
- Program linking (linking automations together)
I can’t over-emphasize enough just how important strategic planning is for your marketing automations! Define goals, a starting point, which channels you’ll need, how you want offers to work, and the frequency at which communications within automations should be sent.
Lastly, look at unique ways to solve common problems — don’t do things just because others did, don’t let email designers hold you up, and don’t build more than you need. The future of automation work will be one where we will need to deal with increasing complexity (technological capabilities and rising customer expectations), but meet these opportunities with increased simplicity, as the artists and creators we are. ◾
You can check out the compilation of my recorded training webinars here to learn more about how to use Predict, Smart Insights, Web Channel, reduce cart abandonments, and more.
Handpicked Related Content:
About the Author
Fred Miller is the U.S. Training Manager at Emarsys. He got his start with training developing curriculum for Apple Retail Stores. He left Apple to pursue a more technical role with a local start-up and soon yearned to return to a training role. Emarsys welcomes Fred with open arms, and he couldn’t be happier. Fred is also founder of a local non-profit and an avid filmmaker and musician.